Participation in clinical trails by community hospitals is a useful means of disseminating state-of-the-art cancer management practices to providers outside university affiliated institutions or major cancer treatment centers. While some cooperative groups have included community affiliates in therapeutic research, few studies have examined the quality of this participation or its effect on cancer care in general for well defined populations, particularly residents of rural areas. The Dartmouth Hitchcock Medical Center (DHMC) currently has a large oncology outreach program that involves direct patient care and continuing education in northern and western New Hampshire and eastern Vermont. We propose now to introduce into this outreach network clinical trials for the common epithelial malignancies. This program will differ from the CCOP, and CGOP programs in that the expertise in the participating hospitals will not come from a pool of dedicated oncologists but primarily from general surgeons who are the rural cancer specialists. Components of the program include workshops and tumor conferences that address which protocols are available and suitable for use in rural hospitals; the procedures for pre-treatment work-up, staging, therapy and follow-up; potential treatment complications; and informed consent and the IRB approval process. DHMC's Protocol Management Service will provide support for the associated data collection activities of the rural hospitals. Participating hospitals will also be provided access to the PDQ and on going consultation services from the clinical staff of the Norris Cotton Cancer Center. The program will be implemented on October 1, 1992 and continue through March 31, 1996. The program's evaluation will compare measures of compliance, data quality, patient accrual and treatment complications with those of DHMC and the cooperative groups offering the protocols (CALGB), RTOG). Separate comparisons will be performed for the one year periods ending on March 31 in 1994, 1995 and 1996. The evaluation will also compare patterns of care (staging procedures, surgery, radiotherapy, chemotherapy, hormonal therapy) during first course of treatment for breast (female), lung and colo-rectal cancer patients diagnosed over a one year period before program implementation (October 1, 1990-September 30, 1991) to those for breast (female), lung, and colo-rectal patients diagnosed two to three years after implementation (October 1, 1994-September 30, 1995). The study population for this part of the evaluation will be restricted to incident cases meeting these criteria who reside in the areas served by the participating New Hampshire hospitals, which will be identified through the New Hampshire State Cancer Registry.